Model efficiency is an important area of model management,
|
|
- Charlene Melton
- 5 years ago
- Views:
Transcription
1
2 Roll Your Own By Bob Crompton Model efficiency is an important area of model management, and model compression is one of the dimensions of such efficiency. Model compression improves efficiency by creating a significantly reduced number of model points compared to a seriatim model. Cluster modeling is a model compression methodology that has been successfully implemented for a number of years. A good introduction to cluster modeling can be found in the article Cluster Analysis: A Spatial Approach to Actuarial Modeling. 1 For simplicity, this article is referred to as the Milliman article. Some actuarial software incorporates cluster modeling; if yours doesn t, this article is for you. SOFTWARE USED The clustering in this article is performed with the open source software R. 2 In addition to the software included in the standard installation, I used two additional packages xlsx and fpc to perform some of the tasks discussed in this article. To install these packages, use the R console commands: install.packages( xlsx, dependencies = TRUE) install.packages( fpc, dependencies = TRUE) Since these packages are not part of the home library, they will need to be added. They can be manually loaded as follows: library(xlsx) library(fpc) CREATING A CLUSTER MODEL I obtained from a colleague an in-force file of universal life (UL) policies. There are five plan types in the file and 1,347 records. The broad steps we need to create a cluster model are: Generate synthetic policy attributes. Apply weighting to the attributes as appropriate. Split data into segments. Import the file of in-force attributes to R. Apply the clustering algorithm. Export results. Reconfigure in-force files for the reduced number of cells and rerun the model. These steps are considered in this article. GENERATE SYNTHETIC POLICY ATTRIBUTES One of the most interesting aspects of the Milliman article is the use of synthetic policy attributes that is, policy attributes that are not found either in the in-force file or are not simple transformations of data found in the in-force file. Quinquennial ages are examples of simple transformations of in-force data. The development of clusters uses attributes associated with projected cash flows in addition to the attributes found in the in-force file. For example, the life/health model used in the Milliman article includes the following synthetic attributes: Present value of proxy profits Present value of proxy profits through 10 projection years Present value of proxy profits through 20 projection years There are three other attributes used in this model, of which two are synthetic. The only native attribute is beginning reserve. It is instructive to review the synthetic attributes for the term life model included in the Milliman article: Beginning reserve Cumulative present value of proxy cash flows Present value of proxy cash flows Years 1 5 Years 6 10 Years Years Years Years Projected death benefits Years 1 5 Years 6 10 Years Years Years Years Projected premiums Years 1 5 Years 6 10 Years Years Years Years DECEMBER 2016 THE MODELING PLATFORM
3 This is 20 attributes, of which only one is native, with all the rest being synthetic. Why are there so many? The complexity of reserving for level term, combined with the mortality patterns during and after the level term period, mean extensive information is required if we want a well-fitting cluster model. The attributes I used for my model were those used in the Milliman article for the traditional life/health model with one exception. I did not include the present value of total proxy profits because my original projections only went for 20 years. The information contained in the early years proxy profits and the later years proxy profits overlaps the information contained in total proxy profits. APPLY WEIGHTS TO ATTRIBUTES Weights adjust the relative importance of the various attributes used for clustering. Weighting affects both the selection of the representative cell for a cluster as well as the cells assigned to the cluster. Consistent with the cluster attributes, the weights I used for my model were those used in the Milliman article for the traditional life/health model. CREATE SEGMENTS The in-force file needs to be split into segments. The Milliman article describes segments as follows: You divide the business into segments, which instructs the program not to map across segment boundaries. Segments might include plan code, issue year, GAAP [generally accepted accounting principles] era or any other dimension of interest. So clustering is applied at the segment level. In my example, the entire file is treated as one segment. If I had used multiple segments, they would have been based on plan code. IMPORTING THE DATA INTO R For the import operation, I am demonstrating how to import from Excel because Excel is ubiquitous. I will demonstrate two possibilities because there is more than one way to skin a cat. (I don t actually know this from personal experience, but generations of folk wisdom attest to the truth of this statement. Who am I to question generations of folk wisdom?) Importing Directly from Excel For the first import operation, I use the read.xlsx function. The read.xlsx function is from the xlsx package. Use the following console command: MyInforce <- read.xlsx( c:/inforce.xlsx, 1) # 2nd parameter indicates which tab to import DECEMBER 2016 THE MODELING PLATFORM 15
4 Roll Your Own MyInforce, the variable containing the output from the read.xlsx operation, is a data frame. Data frames are formatted matrix-like data structures. They are convenient since they can be used in many built-in functions that require matrix input. Importing From a Comma-Separated File For this import operation, I use the read.table function: MyInforce <- read.table( c:/inforce.csv, header = TRUE, sep =, ) The read.table operation yields a data frame, the same as the read.xlsx operation. Comments on Excel vs. CSV Files The read.xlsx function has the obvious benefit of convenience. You are working directly from Excel so you don t have to reformat. In addition, if you put each segment in a separate tab, it is easy to loop through the tabs with the read.xlsx function. DIY clustering is an easy and straightforward process using existing code. However, there are some serious drawbacks to using the read. xlsx approach: It s s-l-o-w! An Excel file with 10,000 records took about 35 minutes to load using read.xlsx. It s memory intensive. My computer was unable to load a file of 25,000 records because the Java back-end to xlsx ran out of memory. Unless all of your segments are less than 2,000 or so records, reading directly from Excel may not be in the cards. Note that the R manual on data import/export has the following warning: The most common R data import/export question seems to be how do I read an Excel spreadsheet. The first piece of advice is to avoid doing so if possible! If you have access to Excel, export the data you want from Excel in tab-delimited or comma-separated form, and use read.delim or read.csv to import it into R. 3 However, if you are committed to using Excel as your data source, the manual contains a description of a few other R packages that provide Excel import capability. Experiment with some or all of these to see if they work any better than the xlsx package. CSV files don t have either of these problems. I was able to load a file of 100,000 records in less than 5 seconds using the read.table function. The only issue I have noted with CSV files is that if you forget to reformat comma-separated numeric values, chaos and darkness will result. APPLY CLUSTERING ALGORITHM Once the data is in R, it is simple to create clusters. There are a number of cluster functions available. I used the pam function because the output contains the information we need to create the clusters. Pam is based on a version of the K-means approach to clustering. The following code is for the cluster model with 13 cells. Note that the data is standardized by setting one of the function parameters shown. Standardization often gives better results when using clustering algorithms. fit <- pam(myinforce, 13, stand = TRUE) The output variable fit is a list containing, among other things, the index number of the representative cells for each cluster in the component id.med, and the cluster assignment of each of the inforce records in the component clustering. So the component id.med is a vector with length equal to the number of clusters (in this instance, 13), and the component clustering is a vector with length equal to the number of in-force records. Table 1 Comparison of Policy Attributes ($1,000s) Attribute Original Clustered Ratio Initial reserve 522, , % Projected first-year premiums 77, , % Projected first-year benefits 38, , % Present value proxy profits, years , , % Present value proxy profits, years (104,775.4) (109,964.9) 105.0% 16 DECEMBER 2016 THE MODELING PLATFORM
5 We pass these back from R using the following export code: write.table(fit$id.med, c:/cluster_index.csv, sep =, ) write.table(fit$id.clustering, c:/cluster_ assignment.csv, sep =, ) It is possible to use the write.xlsx approach to write directly to Excel files; however, the same caveats regarding speed and memory mentioned in the discussion of importing the data will apply to exporting the data as well. RECONFIGURING THE IN-FORCE FILE Once the cluster assignments and the cluster representative cells are determined, reconfiguring the in-force file is straightforward. Static items such as issue age, sex, plan code and similar items are set equal to these items from the representative cell. Dynamic items such as initial reserve, initial fund, amount of insurance and similar items are summed over all the in-force records belonging to the cluster. RESULTS OF THE TEST FILES The results from the test clustering are shown in Table 1. I compare the clustered vs. unclustered results for the total net cash flows and the totals of the synthetic attributes. Observations on Fit Given that this model has a 99 percent compression ratio (13 cells compared to 1,347 in the original model), the fit is reasonable. Model projections for premiums, benefits and distributable earnings are shown in Figure 1. Although the purpose of this article is merely to show how to create cluster models, a brief discussion of how to improve the fit might be helpful. There are two obvious possibilities. The first is to adjust the weighting factors. For example, the weightings for both early and late proxy profits could be increased. This would likely improve the fit for first-year benefits as well as for proxy profits. The second obvious adjustment is to split the model into three or four segments, rather than just one segment. Three segments with four cells each might perform better at fitting the original data since the different UL plans have differing patterns of cash flows and profit emergence. OTHER THINGS TO CONSIDER In addition to simply generating and identifying clusters, there are several other things we can consider as we create cluster models. Figure 1 Model Projections for Premiums, Benefits and Distributable Earnings Original Model Premiums 90,000,000 80,000,000 70,000,000 60,000,000 50,000,000 40,000,000 30,000,000 20,000,000 10,000, ,000,000 70,000,000 60,000,000 50,000,000 40,000,000 30,000,000 20,000,000 10,000, ,000,000 30,000,000 25,000,000 20,000,000 15,000,000 10,000,000 5,000,000 0 Distributable Earnings Original Model Original Model Optimal Number of Clusters What is the optimal number of clusters? If we don t care much about model fit, and are mainly concerned about processing time, then obviously one cluster is optimal. In that case, we simply define the cluster based on the average policy number. ( This is technically known as humor. It is not intended to be taken seriously.) The function pamk in the package fpc estimates the optimal number of clusters based on optimum average silhouette width. A cluster silhouette is a measure of how close each point in one cluster is to points in the neighboring clusters. The further away the points are, the better. DECEMBER 2016 THE MODELING PLATFORM 17
6 Roll Your Own Interestingly, using the pamk function to investigate the optimal number of clusters in the range of two to 100 results in an optimal cluster count of three. That seems to explain how the 99 percent compression model performed as well as it did. Of course, silhouette optimization is not what actuaries are really interested in, so this result does not mean we should automatically choose three clusters. Testing for Sensitivity to Order of Attributes Many K-means algorithms used for clustering seem to be based on the expectation-maximization algorithm. There are some anecdotes that these algorithms are sensitive to the order in which the attributes are presented. The documentation for the pam function claims it is a more robust version of K-means. 4 It is not clear if this means it does not exhibit sensitivity to the order of attributes. I did not bother tracing back to the source reference for the algorithm, but it is something users can easily test. I tested for this sensitivity by running pam with two additional input files that differed only in the column order of the data. Both additional files produced clusters identical to the original in-force file. CONCLUSION As presented here, DIY clustering is an easy and straightforward process using existing code. As my father-in-law used to tell me, It s easy once you know how. In R, once you know how, many things are insanely easy. The difficult part of R is that it is so extensive, finding the right package to do just what you want is a time-consuming task. ENDNOTES Bob Crompton, FSA, MAAA, is a vice president of Actuarial Resources Corporation of Georgia, located in Alpharetta, Ga. He can be reached at bob.crompton@arcga.com. 1 Avi Freedman and Craig Reynolds, Cluster Analysis: A Spatial Approach to Actuarial Modeling, Milliman Research Report, August 2008, uploadedfiles/insight/research/life-rr/cluster-analysis-a-spatial-rr pdf. 2 R Core Team, R: A Language and Environment for Statistical Computing (Vienna: R Foundation for Statistical Computing, 2016), 3 This can be accessed at 4 The pam function is contained in the package cluster: Maechler, M., Rousseeuw, P., Struyf, A., Hubert, M., Hornik, K. (2015). cluster: Cluster Analysis Basics and Extensions. R package version The quickest way to access the documentation for pam is to enter??pam at the command prompt in R. You can also access the documentation at 18 DECEMBER 2016 THE MODELING PLATFORM
Article from: CompAct. July 2009 Issue 32
Article from: CompAct July 2009 Issue 32 Technology Section CompAct ISSUE 32 JULY 2009 Cluster Modeling: A New Technique To Improve Model Efficiency By Avi Freedman and Craig Reynolds 1 Cluster Modeling:
More informationArticle from The Modeling Platform. November 2017 Issue 6
Article from The Modeling Platform November 2017 Issue 6 Actuarial Model Component Design By William Cember and Jeffrey Yoon As managers of risk, most actuaries are tasked with answering questions about
More informationMorningstar Direct Portfolio Analysis & Equity Attribution
Morningstar Direct Portfolio Analysis & Equity Attribution Portfolio Analysis is Morningstar Direct s web-based solution for manager research and due diligence. This tool will provide you with the data,
More informationUser guide for employers not using our system for assessment
For scheme administrators User guide for employers not using our system for assessment Workplace pensions CONTENTS Welcome... 6 Getting started... 8 The dashboard... 9 Import data... 10 How to import a
More informationSession 030 PD - PBR Stochastic Reserve - Challenges and Possible Solutions. Moderator: Sebastien Cimon Gagnon, FSA, CERA, MAAA
Session 030 PD - PBR Stochastic Reserve - Challenges and Possible Solutions Moderator: Sebastien Cimon Gagnon, FSA, CERA, MAAA Presenters: Timothy C. Cardinal, FSA, CERA, MAAA Andrew G. Steenman, FSA,
More informationScheme Management System User guide
Scheme Management System User guide 20-09-2016 1. GETTING STARTED 1.1 - accessing the scheme management system 1.2 converting my Excel file to CSV format 2. ADDING EMPLOYEES TO MY PENSION SCHEME 2.1 Options
More informationIntroduction to Client Online
Introduction to Client Online Trade Finance Guide TradeFinanceNewClientsV2Sept15 Contents Introduction 3 Welcome to your introduction to Client Online 3 If you have any questions 3 Logging In 4 Welcome
More informationPBA Reserve Workshop What Will PBA Mean to You and Your Software? Trevor Howes, FCIA, FSA, MAAA. Agenda. Overview to PBA project
Southeastern Actuaries Conference 2010 Spring Meeting June 16, 2010 PBA Reserve Workshop What Will PBA Mean to You and Your Software? Trevor Howes, FCIA, FSA, MAAA Michael LeBoeuf, FSA, MAAA Agenda Overview
More informationAnalysis of Proposed Principle-Based Approach
Milliman Client Report Analysis of Proposed Principle-Based Approach A review and analysis of case studies submitted by participating companies in response to proposed changes in individual life insurance
More informationChapter 18. Indebtedness
Chapter 18 Indebtedness This Page Left Blank Intentionally CTAS User Manual 18-1 Indebtedness: Introduction The Indebtedness Module is designed to track an entity s indebtedness. By entering the principal
More informationNAIC s Center for Insurance Policy and Research Summit: Exploring Insurers Liabilities
NAIC s Center for Insurance Policy and Research Summit: Exploring Insurers Liabilities Session 3: Life Panel Issues with Internal Modeling Dave Neve, FSA, MAAA, CERA Chairperson, American Academy of Actuaries
More information11/8/2012. Risks and Controls for PBA and PBR VM-G. Acronyms VM-G VM-G. PBA = Principle-Based Approach. PBR = Principle-Based Reserves TOPICS COVERED:
Acronyms PBA = Principle-Based Approach PBR = Principle-Based Reserves Risks and Controls for PBA and PBR Actuaries Club of the Southwest Fall Meeting Houston, TX Rick Farrell FSA, MAAA Director, KPMG
More informationExperience Studies. Southeastern Actuaries Conference. Kevin Pledge FIA, FSA June 2004
Experience Studies Southeastern Actuaries Conference Kevin Pledge FIA, FSA June 2004 Where are we now? Data availability? Studies calculated one at a time? Static reports? Are sales persistency studies
More informationMUNICIPAL REPORTING SYSTEM. SOE Budget (SOE-B) User Guide June 2017
MUNICIPAL REPORTING SYSTEM SOE Budget (SOE-B) User Guide June 2017 Crown copyright, Province of Nova Scotia, 2017 Municipal Reporting System SOE Budget (SOE-B) User Guide Municipal Affairs June 2017 ISBN:
More informationFORECASTING & BUDGETING
FORECASTING & BUDGETING W I T H E X C E L S S O L V E R WHAT IS SOLVER? Solver is an add-in that comes pre-built into Microsoft Excel. Simply put, it allows you to set an objective value which is subject
More informationSession 80 PD, Model Validation Framework and Best Practices. Moderator: Joshua David Dobiac, JD, MS, CAIA
Session 80 PD, Model Validation Framework and Best Practices Moderator: Joshua David Dobiac, JD, MS, CAIA Presenters: James Stuart McClure, FSA, MAAA Zohair A. Motiwalla, FSA, MAAA SOA Antitrust Disclaimer
More informationWashems! Please Note: You must be running Excel 2007 or later in order to use this version of Washems!.
Washems! is a standalone pc-based program designed to calculate disallowed losses resulting from the buying and selling of stocks and options. It was written to satisfy the stringent reporting requirement
More informationPresented to the National Association of Insurance Commissioners Life and Health Actuarial Task Force. San Antonio, TX December 2006
Report on Valuation Effects of a Principle Based Approach ( PBA ) For Accumulation Type Universal Life From the American Academy of Actuaries Life Reserves Work Group Modeling Subgroup Presented to the
More informationModeling Anti-selective Lapse and Optimal Pricing in Individual and Small Group Health Insurance by Andrew Wei
Modeling Anti-selective Lapse and Optimal Pricing in Individual and Small Group Health Insurance by Andrew Wei Andrew Wei, FSA, MAAA, is a vice president, senior modeler with Assurant Health, based in
More informationHow to Use the Savvy Social Security Calculators
appendix a How to Use the Savvy Social Security Calculators The Savvy Social Security Calculators on the enclosed CD utilize Excel spreadsheets to help you run various scenarios when doing Social Security
More informationCCH Fixed Asset Register Quick Start Guide
CCH Fixed Asset Register 2017.1 Quick Start Guide Legal Notice Disclaimer Wolters Kluwer (UK) Limited has made every effort to ensure the accuracy and completeness of these Release Notes. However, Wolters
More informationORIGINALLY APPEARED IN ACTIVE TRADER M AGAZINE
ORIGINALLY APPEARED IN ACTIVE TRADER M AGAZINE FINDING TRADING STRA TEGIES FOR TOUGH MAR KETS (AKA TRADING DIFFICULT MARKETS) BY SUNNY J. HARRIS In order to address the subject of difficult markets, we
More informationSelect Period Mortality Survey
Select Period Mortality Survey March 2014 SPONSORED BY Product Development Section Committee on Life Insurance Research Society of Actuaries PREPARED BY Allen M. Klein, FSA, MAAA Michelle L. Krysiak, FSA,
More informationWeb Extension: Continuous Distributions and Estimating Beta with a Calculator
19878_02W_p001-008.qxd 3/10/06 9:51 AM Page 1 C H A P T E R 2 Web Extension: Continuous Distributions and Estimating Beta with a Calculator This extension explains continuous probability distributions
More informationManaging operational tax risk through technology
Managing operational tax risk through technology EY Africa Tax Conference September 2014 Panel Daryl Blakeway Director Tax Performance Advisory Leader EY South Africa Anthony Davis Director Tax Performance
More informationReserving in the Pressure Cooker (General Insurance TORP Working Party) 18 May William Diffey Laura Hobern Asif John
Reserving in the Pressure Cooker (General Insurance TORP Working Party) 18 May 2018 William Diffey Laura Hobern Asif John Disclaimer The views expressed in this presentation are those of the presenter(s)
More informationState of New Jersey Department of Education Division of Administration & Finance Office of School Finance
2016-2017 State of New Jersey Department of Education Division of Administration & Finance Office of School Finance Audit Summary Online Technical Manual Table of Contents Purpose.. 3 Submission dates....
More informationFinancial Computing with Python
Introduction to Financial Computing with Python Matthieu Mariapragassam Why coding seems so easy? But is actually not Sprezzatura : «It s an art that doesn t seem to be an art» - The Book of the Courtier
More informationMichael Clive Gibson Resume
Nationality: Australian/British Mobile: +44 (0) 7766642218 Michael Clive Gibson Resume www.gibsonactuarial.com Email: michael_c_gibson@hotmail.com Key Strengths Strong technical skills excellent understanding
More informationTechnical User Guide for Advisers
Technical User Guide for Advisers Guide published @ 09/2011 Table of Contents ntroduction... 4 Part 1 How Market Assumptions work in The Tool... 5 Asset Classes... 5 What is the evalue FE CAP:Link TM model?...
More informationChapter 17. Investment Reports
Chapter 17 Investment Reports This Page Left Blank Intentionally CTAS User Manual 17-1 Investment Reports: Introduction There are six reports that you can create and print from the Investment Reports section.
More informationAssessing Solvency by Brute Force is Computationally Tractable
O T Y H E H U N I V E R S I T F G Assessing Solvency by Brute Force is Computationally Tractable (Applying High Performance Computing to Actuarial Calculations) E D I N B U R M.Tucker@epcc.ed.ac.uk Assessing
More informationDECISION SUPPORT Risk handout. Simulating Spreadsheet models
DECISION SUPPORT MODELS @ Risk handout Simulating Spreadsheet models using @RISK 1. Step 1 1.1. Open Excel and @RISK enabling any macros if prompted 1.2. There are four on-line help options available.
More informationVisual Cash Focus - User Tip 39
Visual Cash Focus - User Tip 39 Capital Expenditure and deferred payment options How to input a CAPEX budget and options for deferring payment The Scenario Let s assume that you have created a number of
More informationWashems2019! Basic Required Fields for Each Line in the Input file
Washems2019! is a standalone pc-based program designed to calculate disallowed losses resulting from the buying and selling of stocks and options. It was written to satisfy the stringent reporting requirement
More informationArticle from: Health Section News. October 2004 Issue No. 48
Article from: Health Section News October 2004 Issue No. 48 Read. Think. Write. The Statement of Actuarial Opinion for the Health Annual Statement By Thomas D. Snook and Robert H. Dobson There s more to
More informationIn this chapter: Budgets and Planning Tools. Configure a budget. Report on budget versus actual figures. Export budgets.
Budgets and Planning Tools In this chapter: Configure a budget Report on budget versus actual figures Export budgets Project cash flow Chapter 23 479 Tuesday, September 18, 2007 4:38:14 PM 480 P A R T
More informationAtradius Atrium. July version 3.0. Atradius Atrium. User manual. Atradius Atrium - User Manual Version 3.0
July 2018 - version 3.0 User manual 1 - User Manual Version 3.0 Drive your business forward with powerful, easy-to-use credit management tools is the Atradius online platform, which offers you one place
More informationUsing FastCensus for Plan Sponsors
Using FastCensus for Plan Sponsors FastCensus is a secure, online tool for Plan Sponsors to access, edit, validate and submit census data to their Third Party Administrator for the purposes of year-end
More informationA Guide to Month-end & Year-end Accounting
A Guide to Month-end & Year-end Accounting Version 2015.2 Page 1 Contents Structure of Xebra Accounting After you have reconciled all your Bank Accounts: Reports - General Ledger Balance - Accrual Basis
More informationFull Monte. Looking at your project through rose-colored glasses? Let s get real.
Realistic plans for project success. Looking at your project through rose-colored glasses? Let s get real. Full Monte Cost and schedule risk analysis add-in for Microsoft Project that graphically displays
More informationPCGENESIS FINANCIAL ACCOUNTING AND REPORTING (FAR) SYSTEM OPERATIONS GUIDE
PCGENESIS FINANCIAL ACCOUNTING AND REPORTING (FAR) SYSTEM OPERATIONS GUIDE 4/1/2011 Section A: Budget Account Master Processing, V2.1 Revision History Date Version Description Author 04/1/2011 2.1 11.01.00
More informationTOOLBOX FUNCTION: Import G/L Budget Entries
TOOLBOX FUNCTION: Document Ref: TBO-005 Date: Jun 12 2002, rev. 05/02/05, 07/05/07, 01/22/14 Document Version: 0.4 Modules Affected: Earliest available version of COINS: Documentation Updated: General
More informationUsing the Burn Rate Tool
Using the Burn Rate Tool Mini-Webinar #3 Page 1 Hi, I'm Jeanine, and these are my colleagues Ruth, Dan and Jitesh. We have come together to help each other, and you, better understand budgeting and the
More informationIssue Selection The ideal issue to trade
6 Issue Selection The ideal issue to trade has several characteristics: 1. There is enough data to allow us to model its behavior. 2. The price is reasonable throughout its history. 3. There is sufficient
More informationColumbia University. Actuarial Science Integrated Project. Group Six
Columbia University Actuarial Science Integrated Project Group Six Variability of Universal Life Cash Flows under Higher Risk Investment Strategies Mentor: Students: Bob Stein, CPA, FSA Abhishek Tayal
More informationMaster User Manual. Last Updated: August, Released concurrently with CDM v.1.0
Master User Manual Last Updated: August, 2010 Released concurrently with CDM v.1.0 All information in this manual referring to individuals or organizations (names, addresses, company names, telephone numbers,
More informationTutorial. Morningstar DirectSM. Quick Start Guide
April 2008 Software Tutorial Morningstar DirectSM Quick Start Guide Table of Contents Quick Start Guide Getting Started with Morningstar Direct Defining an Investment Lineup or Watch List Generating a
More informationRe: ASB Comments Comments on Second Exposure Draft of the Modeling ASOP
March 1, 2015 Modeling (Second Exposure) Actuarial Standards Board 1850 M Street NW, Suite 300 Washington, DC 20036 Re: ASB Comments Comments on Second Exposure Draft of the Modeling ASOP Members of the
More informationPCGENESIS FINANCIAL ACCOUNTING AND REPORTING (FAR) SYSTEM OPERATIONS GUIDE
PCGENESIS FINANCIAL ACCOUNTING AND REPORTING (FAR) SYSTEM OPERATIONS GUIDE 9/3/2015 Section A: Budget Account Master Processing, V2.5 Revision History Date Version Description Author 9/3/2015 2.5 15.03.00
More informationErrors in Operational Spreadsheets: A Review of the State of the Art
Errors in Operational Spreadsheets: A Review of the State of the Art Abstract Spreadsheets are thought to be highly prone to errors and misuse. In some documented instances, spreadsheet errors have cost
More informationScenario and Cell Model Reduction
A Public Policy Practice note Scenario and Cell Model Reduction September 2010 American Academy of Actuaries Modeling Efficiency Work Group A PUBLIC POLICY PRACTICE NOTE Scenario and Cell Model Reduction
More informationVisual Cash Focus - User Tip 39
Visual Cash Focus - User Tip 39 Capital Expenditure, Deferred Payment and Financing options How to input a CAPEX budget and options for deferring and financing payments The Scenario Let s assume that you
More informationå Follow these steps to delete a list: å To rename a list: Maintaining your lists
Maintaining your lists TradingExpert Pro provides a number of functions for maintaining the data contained in your Group/Sector List and all other lists that you have created. This section lists the data
More informationTREASURER/SECRETARY S DUTIES
Updated: October, 2009 TREASURER/SECRETARY S DUTIES Ref: DEYC By-Laws Article IV, Section 1 and Article V, Section 6 The Treasurer/Secretary is elected at the Annual Meeting of the Club. As an officer
More informationFlight Ascend Values Analyzer User Guide. Introduction. Values Home Page. Navigation around the product
Flight Ascend Values Analyzer User Guide Introduction Flight Ascend Values Analyzer (FAVA) is a new platform designed to replace ASO Values and become the unified platform for accessing all FlightGlobal
More informationMultiple regression - a brief introduction
Multiple regression - a brief introduction Multiple regression is an extension to regular (simple) regression. Instead of one X, we now have several. Suppose, for example, that you are trying to predict
More informationFRx FORECASTER FRx SOFTWARE CORPORATION
FRx FORECASTER FRx SOFTWARE CORPORATION Photo: PhotoDisc FRx Forecaster It s about control. Today s dynamic business environment requires flexible budget development and fast, easy revision capabilities.
More informationFRTB. NMRF Aggregation Proposal
FRTB NMRF Aggregation Proposal June 2018 1 Agenda 1. Proposal on NMRF aggregation 1.1. On the ability to prove correlation assumptions 1.2. On the ability to assess correlation ranges 1.3. How a calculation
More informationFREQUENTLY ASKED QUESTIONS
General Questions: Questions 1. How should store sites be named? 2. How do I get help? 3. How to request consultant/vendor access? 4. How to request FBO Vendor access? 5. How do I delete a project? Responses
More informationMun-Ease News. A Behind the Scenes Look At Release Release 2000 Architecture. New Database Format. A New Tutorials Volume / Examples Database
Mun-Ease News www.mun-ease.com 08/15/2000 A Behind the Scenes Look At Release 2000 Release 2000 Architecture We first began developing Release 2000 in May of 1999. As you may know, Mun-Ease is written
More informationMay Link Richardson, CERA, FSA, MAAA, Chairperson
Recommended Approach for Updating Regulatory Risk-Based Capital Requirements for Interest Rate Risk for Fixed Annuities and Single Premium Life Insurance (C-3 Phase I) Presented by the American Academy
More informationR For Actuaries: What, Why and Where? 26 th October 2017
R For Actuaries: What, Why and Where? 26 th October 2017 Disclaimer The views expressed in this presentation are those of the presenters and not necessarily of the Society of Actuaries in Ireland What
More informationIntroduction to Client Online
Introduction to Client Online Bibby Factors International Guide 1 InternationalFactoringNewClientBibbyUKopsSept15 Introduction 3 Logging In 5 Welcome Screen 6 Navigation 7 Viewing Your Account 9 Invoice
More informationFrequency Distributions
Frequency Distributions January 8, 2018 Contents Frequency histograms Relative Frequency Histograms Cumulative Frequency Graph Frequency Histograms in R Using the Cumulative Frequency Graph to Estimate
More informationExplaining Your Financial Results Attribution Analysis and Forecasting Using Replicated Stratified Sampling
Insights October 2012 Financial Modeling Explaining Your Financial Results Attribution Analysis and Forecasting Using Replicated Stratified Sampling Delivering an effective message is only possible when
More informationFinance, Management & Operations
Finance, Management & Operations Successful Partnering with Actuarial Betsy Kadanoff, Senior Director John Hancock Lisa White, Director Bankers Life and Casualty Company March 2015 Successful Partnering
More informationThis document describes version 1.1 of the Flexible Quota System.
POLAR BEAR FLEXIBLE QUOTA SYSTEM This document describes version 1.1 of the Flexible Quota System. INTRODUCTION The flexible quota system for polar bears is assumes that the annual maximum sustainable
More informationIntroduction to Client Online
Introduction to Client Online Construction Finance Guide ConstructionFinanceNewClientsV2Sept15 Contents Introduction 3 Welcome to your introduction to Client Online 3 If you have any questions 3 Logging
More informationActuarial Concepts 101. Presented By: Jason L. Franken, FSA, EA, MAAA
Actuarial Concepts 101 Presented By: Jason L. Franken, FSA, EA, MAAA ROLE OF THE ACTUARY Determine the Timing and Pattern of Annual Contributions Help Maintain the Health of the Pension Fund Ensure that
More informationFINANCIAL OPTIMIZATION
FINANCIAL OPTIMIZATION Lecture 2: Linear Programming Philip H. Dybvig Washington University Saint Louis, Missouri Copyright c Philip H. Dybvig 2008 Choose x to minimize c x subject to ( i E)a i x = b i,
More informationHertha Longo, CSA Matt Wade
Hertha Longo, CSA Matt Wade Census and financial forecasting tool to assist religious institutes in decision-making and planning for the future Available on CD for purchase by religious institutes only
More informationBudget Forecast Return 2016 to 2017 (to be completed by all academies)
For the Attention of the SIMS FMS6 Operator SIMS FMS6 Academies USER BULLETIN No.A28 2016/17 Budget Forecast Return ---------- Financial Services for Schools Helpline Tel: 01992 555753 Fax: 01992 555727
More informationHPE Project and Portfolio Management Center
HPE Project and Portfolio Management Center Software Version: 9.41 Financial Management User's Guide Go to HELP CENTER ONLINE http://ppm-help.saas.hpe.com Document Release Date: March 2017 Software Release
More informationPresenter: And Paul, you've been quite vocal on the inadequacies of the SRRI calculation.
Morningstar - KIID Key Investor Information Document - KIID Paul Kaplan, Jeff Strazis & Neil Simmonds Presenter: I'm joined now by Neil Simmonds, Partner at Simmons & Simmons, Dr Paul Kaplan, Director
More informationBetter decision making under uncertain conditions using Monte Carlo Simulation
IBM Software Business Analytics IBM SPSS Statistics Better decision making under uncertain conditions using Monte Carlo Simulation Monte Carlo simulation and risk analysis techniques in IBM SPSS Statistics
More information36106 Managerial Decision Modeling Monte Carlo Simulation in Excel: Part IV
36106 Managerial Decision Modeling Monte Carlo Simulation in Excel: Part IV Kipp Martin University of Chicago Booth School of Business November 29, 2017 Reading and Excel Files 2 Reading: Handout: Optimal
More informationDUALITY AND SENSITIVITY ANALYSIS
DUALITY AND SENSITIVITY ANALYSIS Understanding Duality No learning of Linear Programming is complete unless we learn the concept of Duality in linear programming. It is impossible to separate the linear
More informationAnnouncing Pro Tools V2.0 March 28, 2015
Announcing Pro Tools V2.0 March 28, 2015 A letter from the CEO.. Greetings OmniVest subscribers! Spring is in the air and it s the perfect time for me to update you on our magnificent OmniVest Professional
More informationRegions Integrated Receivables User Guide
Regions Integrated Receivables User Guide 01/22/18 TABLE OF CONTENTS GENERAL GUIDELINES... 3 DASHBOARD - OVERVIEW... 3 Information included on the Summary Page... 4 Detail Payment Screen... 5 ACH Dashboard...
More informationRTD Documentation. =RTD( progid, server, [Field1], [Field2],...)
RIT User Guide Build 1.00 RTD Documentation The RTD function in Excel can retrieve real-time data from a program, such as the RIT Client. In general, the syntax for an RTD command is: =RTD( progid, server,
More informationAvailable Format (See descriptions below) Comma Delimited. Fixed Space Text PDF (Print Report) Detail X X X X Summary X X Contribution Type*
14. Trust reports Reports: Trust reports You can request and download trust reports for custom date ranges directly from the Plan Sponsor website. Detail, Summary and Contribution type reports are available.
More informationConsumer Research: overdrafts and APR. Technical Report. December 2018
Consumer Research: overdrafts and APR. Technical Report December 2018 TECHNICAL REPORT 1. Introduction This technical report relates to research on overdrafts and APR published in the technical annex to
More informationChapter 15: Dynamic Programming
Chapter 15: Dynamic Programming Dynamic programming is a general approach to making a sequence of interrelated decisions in an optimum way. While we can describe the general characteristics, the details
More information4. Basic distributions with R
4. Basic distributions with R CA200 (based on the book by Prof. Jane M. Horgan) 1 Discrete distributions: Binomial distribution Def: Conditions: 1. An experiment consists of n repeated trials 2. Each trial
More informationArticle from. The Actuary. October/November 2015 Issue 5
Article from The Actuary October/November 2015 Issue 5 FEATURE PREDICTIVE ANALYTICS THE USE OF PREDICTIVE ANALYTICS IN THE DEVELOPMENT OF EXPERIENCE STUDIES Recently, predictive analytics has drawn a lot
More informationGlobal Liquidity Fund service user guide
Global Liquidity Fund service user guide Contents Page 1 Welcome 1 2 Using the Global Liquidity Fund service for the first time 2 3 Account maintenance 4 4 Investment Reports 7 5 Create New Investment
More informationProject Manager Workbench in SAP ECC or S4 HANA : 360 Financial Management of projects directly in SAP
Project Manager Workbench in SAP ECC or S4 HANA : 360 Financial Management of projects directly in SAP Business Challenges Symptoms: You have implemented SAP Project System, but you use it as a cost collector
More informationWinTen² Budget Management
Budget Management Preliminary User Manual User Manual Edition: 4/13/2005 Your inside track for making your job easier! Tenmast Software 132 Venture Court, Suite 1 Lexington, KY 40511 www.tenmast.com Support:
More informationTo assist recordkeepers in this task, Mid Atlantic has developed the Investment Toolkit. The Toolkit provides the following services:
Overview The new fee disclosure rules have created new responsibilities for plan sponsors and plan service providers. While overall, these new rules should help even the playing field for unbundled or
More informationCSV Import Instructions
CSV Import Instructions The CSV Import utility allows a user to import model data from a prepared CSV excel file into the Foresight software. Unlike other import functions in Foresight, you will not create
More informationFATCA Administration and Configuration Guide. Release April 2015
FATCA Administration and Configuration Guide Release 6.2.5 April 2015 FATCA Administration and Configuration Guide Release 6.2.5 April 2015 Part Number: E62969_14 Oracle Financial Services Software, Inc.
More informationJacob: What data do we use? Do we compile paid loss triangles for a line of business?
PROJECT TEMPLATES FOR REGRESSION ANALYSIS APPLIED TO LOSS RESERVING BACKGROUND ON PAID LOSS TRIANGLES (The attached PDF file has better formatting.) {The paid loss triangle helps you! distinguish between
More informationThe Dynamic Cross-sectional Microsimulation Model MOSART
Third General Conference of the International Microsimulation Association Stockholm, June 8-10, 2011 The Dynamic Cross-sectional Microsimulation Model MOSART Dennis Fredriksen, Pål Knudsen and Nils Martin
More informationFinancial Wellness Essay Collection
Article from Financial Wellness Essay Collection 2017 Call for Essays Copyright 2017 Society of Actuaries. All rights reserved. Using Sound Actuarial Principles to Enhance Financial Well-Being Ken Steiner
More informationModel Governance: Is YOUR Company There Yet? Past, Present and Future of Model Governance. Moderator: Ronald J. Harasym, FSA, CERA, FCIA, MAAA
Model Governance: Is YOUR Company There Yet? Past, Present and Future of Model Governance Moderator: Ronald J. Harasym, FSA, CERA,, FCIA, MAAA Presenters: Dave Czernicki, FSA, MAAA Ronald J. Harasym, FSA,
More informationNSP-41. The Wealth Building Strategy. The ONLY Trading System with a One-Year Money Back Guarantee! Limited to 300 Copies!
NSP-41 The Wealth Building Strategy Limited to 300 Copies! The ONLY Trading System with a One-Year Money Back Guarantee! NirvanaSystems For the Trader Serious about Building Wealth The Path to Trading
More information<Partner Name> <Partner Product> RSA ARCHER GRC Platform Implementation Guide. 6.3
RSA ARCHER GRC Platform Implementation Guide Palisade Jeffrey Carlson, RSA Partner Engineering Last Modified: 12/21/2016 Solution Summary Palisade @RISK is risk and decision
More informationTop 7 IFRS Mistakes. That You Should Avoid. Silvia of IFRSbox.com
Top 7 IFRS Mistakes That You Should Avoid Learn how to avoid these mistakes so you won t be penalized or create an accounting scandal at your company. Silvia of IFRSbox.com Why Top 7 Mistakes That You
More informationTrading Diary Manual. Introduction
Trading Diary Manual Introduction Welcome, and congratulations! You ve made a wise choice by purchasing this software, and if you commit to using it regularly and consistently you will not be able but
More information